Adaptive learning focuses on environments that are responsive to learners as individuals.
The National Research Council's
How People Learn (HPL) framework
refers to this quality as being
learner centered, one of four fundamental aspects of
effective learning environments it emphasizes.
Adaptivity relies on recognizing individual learner differences and understanding
how different learner situations can influence progress towards successful outcomes.
Adaptations encode this understanding, altering flow or materials to
better scaffold each learner.
Over 35 years of research have
been dedicated to applying intelligent systems techniques to creating
technology-supported adaptive learning environments, represented today by
intelligent tutoring systems
and adaptive hypermedia.
These technologies focus on adaptation principally as a means of
guiding learners through a corpus of instruction,
typically based on an individual learner's differences from domain experts.
While our work is related to this inquiry,
we focus on fine-grained, lightweight adaptations
situated within learning activities and materials.
Our motivation is to enable educators to incrementally introduce adaptive features into
their learning designs
by making simpler capabilities for representation, interaction, and reasoning more accessible.
Our work is also related to the current generation of
e-learning technologies and
standards,
although there are significant
differences in context and focus. Addressing use
in the context of schools, we focus on blended learning more than distance learning, as
evidenced by technology features for coordinating
classroom-based and outside-class learning. As a particular
emphasis, we promote a reflective practice of design by educators
through the use of embedded formative assessment, together with unobtrusive observations of learners,
as vehicles for improving the responsiveness and efficacy of learning designs over time.
Such improvements can be pursued individually by educators or cooperatively for designs shared
within a community.
This site provides information about the technologies we have developed to support adaptive online learning
and their use in blended learning environments.
It provides descriptions of our visual authoring technology, called CAPE, and our
adaptive web-based learning platform eLMS.
Together these technologies comprise an infrastructure with novel capabilities for
creating, enacting, and evolving learner-centered designs.
Examples of such designs, created by educators participating in the
VaNTH ERC,
are available online at the
VaNTH Portal.
A description of this dissemination vehicle and its integration with CAPE and eLMS is provided on this site.
Howard, L., Johnson, J., & Neitzel, C.
Examining learner control in a structured inquiry cycle using process mining.
Proceedings of the 3rd International Conference of Educational Data Mining (EDM'2010). pp. 71-80.
Joeng, H., Biswas, G., Johnson, J. & Howard, L.
Analysis of productive learning behaviors in a structured inquiry cycle using hidden Markov models.
Proceedings of the 3rd International Conference of Educational Data Mining (EDM'2010). pp. 81-90.
Howard, L., Johnson, J., and Neitzel, C.
Reflecting on online learning designs using observed behavior.
Proceedings of the Fifteenth Annual Conference on Innovation and Technology in Computer Science Education (ITiCSE '10). ACM, New York, NY, USA, 179-183.
Howard, L., Johnson, J., Pap, G., Pence, K., Juracz, L.,
Affordances and Form: Applying Lessons from Informal Learning to Formal Learning on the Web.
Proceedings of the International Conference on Engineering Education. Pecs-Budapest, Hungary, July, 2008.
Roselli, R.J., Gilbert, S.B., Howard, L., Blessing, S.B., Raut A., and Pandian P.,
Integration of an Intelligent Tutoring System with a Web-based Authoring System to Develop Online Homework Assignments with Formative Feedback.
Proceedings of the American Society for Engineering Education Annual Conference, Pittsburgh, PA, USA, July, 2008.
Imsand, E., Howard, L., Pence, K., and Dasgupta, D., Online Information Security Education through Anchored Instruction.
Proceedings of the 12th Annual Colloquium for Information Systems Security Education, Dallas, TX, USA, June, 2008.
Roselli, R.J., Howard, L. and Pap, G., Assessment of Diagnostics and Feedback for Online Homework Assignments.
Annual BMES Conference, Los Angeles, CA (CD ROM), 2007.
Harris, T. R., Howard, L., and Linsenmeier, R., Design of a Repository for Bioengineering Educational Materials.
Biomedical Engineering Society (BMES) Annual Conference, Chicago, Illinios, October, 2006.
Howard, L., Remenyi, Z. and Pap, G., Adaptive blended learning environments.
International Conference on Engineering Education, Session T3K, July, 2006.
Debelak, K., Howard, L., Xue, Y., Lee, C. and Sztipanovits, J., Introducing security in a chemical engineering
design course using adaptive online learning. International Conference on Engineering Education, Session T1H, July, 2006.
Roselli, R.J., Howard, L., and Brophy S., Integration of formative assessment into online engineering assignments.
Computers in Education, Vol. 16, No. 4 (2006) 8-17.
Roselli, R.J., Howard, L. and Brophy, S., A computer-based free body diagram assistant.
Computer Applications in Engineering Education, Vol. 14, Issue 4 (2006) 281-290.
Roselli, R.J. and Howard, L., Facilitating the design of diagnostic learning modules with CAPE.
Proceedings of the American Society of Engineering Education Annual Conference, Session 3409, June, 2006.
Sztipanovits, J., Biswas, G., Frampton, K., Gokhale, A., Howard, L., Karsai, G., Koo, T. J., Koutsoukos, X., and Schmidt, D.C.,
Introducing embedded software and systems education and advanced learning technology in an engineering curriculum.
Trans. on Embedded Computing Sys. 4, 3 (Aug., 2005), 549-568.
Howard, L., Adaptive learning technologies for bioengineering education, IEEE Engineering in Medicine and
Biology Magazine, 22, 4, pp. 58-65, July, 2003.
Roselli, R.J., Howard, L., et al. Integration of an interactive Free Body Diagram assistant with a courseware authoring package and an experimental learning management system.
Proceedings of the American Society for Engineering Education (CD-ROM DEStech Publications) Session 2793: 10 pages (2003).
Rothney, M.P., Roselli, R.J., and Howard, L., Creation of an online vector addition tutorial: Exploring the advantages of providing diagnostic, multilevel feedback in basic skills remediation,
ASEE Annual Conference, (CD-ROM DEStech Publications) Session 2793: 7 pages, 2003.
Roselli, R.J. and Howard, L. Development of online homework problems that provide instant feedback and remediation to students.
Annual BMES Conference, Nashville, TN (CD ROM, Omnipress), 2003.
Howard, L., CAPE: A visual language for courseware authoring. Second Workshop on Domain-Specific Visual Languages,
Seattle, WA, November 4, 2002.
A largely non-technical overview of our work is available in a
paper appearing in the
IEEE Engineering in
Medicine and Biology Magazine (22:4). Our current year
project summary, taken from the
VaNTH ERC Annual Report,
provides additional details of our goals, status, and future plans.
CAPE's online documentation
is also a valuable source of information about the technologies.
Events
Below are public meetings where our technologies and learning applications built with them will be presented.
June 2-4, 2008 Dallas, Texas
Colloquium for Information Systems Security Education
July 22-25, 2008 Pittsburgh, Pennsylvania
American Society for Engineering Education Annual Conference
July 27-31, 2008 Pecs-Budapest, Hungary
International Conference on Engineering Education
CAPE and eLMS are freely available to VaNTH and TRUST participants from primary and affiliated institutions.
The technologies can be made available to others
by agreement with the VaNTH ERC. Use the contact information below for inquiries.
Contact:
Larry Howard
Sr. Research Scientist
Institute for Software Integrated Systems
Vanderbilt University
(615.343.7447)
For technical support, CAPE and eLMS developers can be contacted at the following e-mail addresses:
CAPE
(cape_developers@isis.vanderbilt.edu)
eLMS
(elms_developers@isis.vanderbilt.edu)
Mailing list have been established for CAPE and eLMS users. You can subscribe at:
CAPE
(http://list.isis.vanderbilt.edu/mailman/listinfo/cape-users)
eLMS
(http://list.isis.vanderbilt.edu/mailman/listinfo/elms-users)
Our motto,
Eruditio pro totus (
Learning for all), captures for us what is essential about adaptive learning.
First is that educators should view themselves as learners,
with a responsibility to consider poor performance, or failure, by individual learners as potentially
a reflection on the instruction itself.
Openness to this possibility is necessary to motivate inquiry by educators to understand failure,
and from this understanding the consideration of opportunities to reduce it.
Second is that learners are individuals, and, despite all attempts to achieve a uniform preparation for the instruction they
are to receive, they will nonetheless exhibit differences in prior knowledge and skills, differences in motiviations in pursuing
the instruction arising from individual interests, differences in receptivity to selected instructional methods,
and, therefore, different experiences when pursuing the given instruction.
Often times learners can accommodate their differences from assumptions embodied in the instruction's design through
their own adaptivity.
But the responsibility for adaptation should not be the learner's alone.
While adaptive learning is not made possible by technology, we see roles for technology to play.
Technology can enable a richer understanding of how learners progress through learning activities,
by virtue of greater instrumentation of the learning environment.
And technology can enable this understanding
to be encoded as adaptations within the design of learning experiences that meet special needs of
individual learners.
The primary aim of our research has been to facilitate these possible roles.
Our adaptive learning infrastructure consists of two primary technologies:
 |
The Courseware Authoring and Packaging Environment |
 |
The experimental Learning Management System |
The diagram below shows some of the relationships between CAPE and eLMS.
The Flash version can be animated using the controls in the upper right corner.
The infrastructure reflects a separation of concerns between learning materials and how they
are used to create a learning experience.
The design specifications created with CAPE are basically "instructions" to some delivery machinery
embedded in an eLMS server for how to enact an experience with a learner, potentially performed
over multiple sessions.
The materials themselves are created with conventional web authoring tools or else repurposed from
existing web content.
What is essential in this separation is that traditional link-based navigation among content elements is
replaced by the learning design itself.
This allows reasoning performed when enacting the design to change when materials are made available to the learner
and how they are presented.
Another way of looking at the relationships between CAPE and eLMS is by means of a
concept map.
The concept map below summarizes various details about the technologies and the courseware they create and deliver.
If the map's text is too small, you can view it in a
separate window.
The CAPE authoring environment supports an
enactable
design representation for on-line learning designs involving
static, interactive, and dynamic content elements
created with conventional web authoring tools and within CAPE itself.
The designs specify when, or under what circumstances, content elements are presented to a learner
during the course of a learning experience.
Interactive elements can elicit information from a learner, and the outcomes
are available immediately to adaptations incorporated into designs.
A data modeling facility enables capturing facts, including data defined abstractly by
expression, for use in realizing adaptation schemes.
Simple sequencing constructs can be extended with computational components
for more advanced reasoning.
Sequencing Model in the CAPE Authoring Environment
CAPE designs involve other kinds of specifications in addition to sequencing models.
These include the statement of
learning objectives and their association with content knowledge
represented by curricular taxonomies, as well as tagging with
community-specific and standards-based
metadata.
Such specifications play no direct role in design enactment, but are used to communicate the
original designer's intentions to other authors and to provide additional descriptions of elements
and resources for use by other tools.
The design representation used by CAPE is a domain-specific visual language,
where hierarchically organized icons and connections represent concepts and relationships
in the language, respectively, and attributes uniquely characterize occurences.
The choice of a visual language for CAPE reflects our interest in a representation that is
both expressive and easily constructed.
The interface style for such languages, consisting predominantly of drag, drop, interconnect, and specify operations,
offers many affordances over textual representations.
As a desktop application, CAPE lacks the convenience of web-based authoring tools, but enjoys
distinct advantages over forms-based authoring, especially in terms of scalability.
Principal features of the CAPE design representation are presented in the concept map below.
If the map's text is too small, you can view it in
a separate window.
CAPE supports both elaborative (top-down) and integrative (bottom-up) approaches to design.
Rapid prototyping of adaptation schemes can be performed prior to content development.
Existing content and design elements can be readily incorporated into new designs.
The environment supports design-time adaptation by providing
abstraction facilities that can be used to capture invariants
among families of designs and elements as
instructional design patterns.
While CAPE—as a general-purpose design tool—is pedagogically neutral, these design abstractions
can be used to scaffold particular learning strategies that can then be
shared with other authors through an integrated web-based design repository.
CAPE provides a set of extension components that assist the author in
creating, previewing, and packaging designs.
An event-based agent continuously monitors the author's actions
looking for opportunities to provide time-saving assistance.
An online learning component makes CAPE-authored tutorials directly available
within the design environment to support just-in-time learning.
The environment can be extended with wizards that automate complex or repetitive actions.
A design previewing component is complemented with a web-based debugger.
Content and computational elements can be interchanged with traditional development tools.
Completed designs can be directly uploaded to the delivery platform for subsequent assignment to learners.
These and other capabilities of the CAPE design environment are summarized in the concept map below.
If the map's text is too small, you can view it in
a separate window.
CAPE is built on open source technologies from
ISIS—particularly, the
Generic Modeling
Environment (GME) and Meta-GME—and uses
the open source Python dynamic programming language
for realizing its extension components and for computational aspects of CAPE designs.
eLMS is an adaptive learning platform that supports interoperation using web services,
both in conjunction with enacting courseware designs
and in managing domain-specific objects, such as classes, users, and courseware.
The heart of the eLMS platform is a model-based delivery engine that enacts
learning designs authored with the CAPE design environment.
The platform automatically captures detailed instrumentation of these design enactments, and
additional instrumentation—to support grading using custom rubrics, for example—can
be incorporated into courseware designs with CAPE.
The resulting delivery records can be
queried by instructors and authors using an integrated data mining facility.
These capabilities enable an intimate understanding of what learners actually do with on-line learning
experiences, which is essential to making incremental improvements over time.
Courseware Delivery with eLMS
Profiles can be used to collect information about learners, classes, and courseware resulting from
design enactments.
Courseware profiles can be used to collect statistics about the use of a particular
learning design, whether this use occurs between semesters at a single institution or
across multiple institutions.
Learner profiles can collect information elicited from a learner during an earlier courseware for use
as part of an adaptation scheme in a subsequent courseware.
Class profiles can create digests of information from assignments performed outside class for use in scaffolding
classroom-based learning experiences.
Shared state is a sufficient condition for enabling many kinds of interactions between learners, and between learners
and an instructor or teaching assistants.
The Courseware and Class profiles provide such shared state,
and CAPE-authored designs can leverage these capabilities to create many kinds of collaborative services
embedded within eLMS courseware.
An example is support for asynchronous dialoging between an instructor and the students in a class.
We provide a demonstration of using
the Class profile to support such interactions.
Another example of using the shared state provided by profiles is the implementation of experimental research designs
to evaluate courseware design alternatives, for example.
One possible research design involves randomly dividing the students in a class into groups and providing
each group a different "treatment", one of which typically serves as a control.
We provide a demonstration of using
the Class profile to support such research designs.
eLMS allows learners to review materials and activities across
multiple sessions, to take private notes that can be exported from the learning environment, and to access
context-sensitive help resources provided by learning designs.
eLMS instructors and teaching assistants can manage the rosters of classes and make courseware assignments to a class or
to individuals in the class.
The status of learners completing assignments can be monitored,
learners can be selectively released from synchronization points defined by learning designs,
and instructors can replay assignments with learners during face-to-face meetings.
Courseware revisions uploaded by authors are differentially versioned
to avoid disruption of in-progress enactments with learners.
While eLMS can be employed directly to manage
the use of CAPE-authored designs by classes of learners,
it can also be transparently embedded into other learning platforms.
We have developed a building block integration of eLMS with
the Blackboard Learning System,
as an example of a custom integration.
With this plug-in,
instructors can assign eLMS courseware to their learners just as any
other kind of Blackboard assignment.
Documentation
for this embedding solution is available online.
As a somewhat weaker—but more broadly applicable—form of embedding,
we support packaging eLMS courseware using the SCORM standard,
thereby enabling delivery from a standards-compliant learning platform.
This approach to integration is similar to the SCORM platform delivering material
from an external content repository.
With this solution, eLMS courseware can be transparently delivered from a variety of commercial learning platforms, such
as WebCT, as well as non-commercial platforms, such as
Moodle and Sakai.
When there is no local learning platform, we provide one additional option for delivering eLMS courseware called Quota Links.
These are essentially limited-use URLs that include a weak form of authentication with courseware launching.
Users accessing such links are asked to identify themselves and choose a passcode that can be used to re-access the courseware
at a later time, allowing courseware delivery over multiple browser sessions.
Delivery records from such accesses can be retrieved by the eLMS user who creates the Quota Link.
The effect is to enable a class to be dynamically formed as learners access a Quota Link for courseware delivery.
As the name suggests, Quota Links allow only a limited number of deliveries of the associated courseware to be created.
These and other capabilities of the eLMS learning platform are summarized in the concept map below.
If you wish, you can view it in a separate window.
eLMS is built on open source technologies, including the Zope web application server
and Apache.
All eLMS servers are currently deployed on the secure OpenBSD
operating system and are securely accessed by web-services and browser clients using SSL connections.
The
VaNTH Portal is a web-accessible repository serving as a dissemination vehicle for the VaNTH ERC.
It makes classroom-based, outside class, and blended learning materials and experiences designed by VaNTH educators
available to the larger bioengieering educational community.
The portal is profile-based.
Profiles scaffold learning materials with sets of resources that contribute to their reusability, such as instructor guides and
research papers that report on instructional designs and assessments of efficacy.
Each profile is classified with a set of metadata that describes such things as educational level, development status,
and delivery mode.
Profiles provide courseware at varying granularities, from individual modules to entire courses.
Courseware Profile
Profiles can be created and modified by authors through the web.
Workflows support an editorial process that can be tailored to support institutional or domain-based reviewers.
Profiles approved by the editorial process are then made available to visitors.
Requests to access the learning materials provided by profiles involve making an agreement with the ERC, a process automated by the portal.
These requests can be individually tracked by the originating educator through the portal or by e-mail notifications.
The portal complements simple keyword-based searching of courseware profiles with an advanced metadata-based search interface based on the
retrieval by reformulation paradigm.
This interface helps visitors during the search process by providing feedback on the current state of their search in terms of matching
metadata from available profiles.
This feedback ensures that metadata-based queries will always match at least one profile on the portal.
The portal is integrated with other VaNTH learning technologies.
Resources for adaptive online learning, delivered by an eLMS server, can be directly launched from the portal to support evaluation by
visitors of the site.
Repository-based CAPE courseware designs can be bundled with online learning resources in profiles,
and these designs can be inspected and adapted by educators that install the CAPE authoring environment.
CAPE supports publishing profiles to the portal for online learning designs created or adapted by authors.
These forms of integration provide much greater capabilites for adaptive reuse than web-based tools that support simple bundling and sequencing.
These and other capabilities of the VaNTH Portal are summarized in the concept map below.
If the map's text is too small, you can view it in
a separate window.
The VaNTH Portal is built on the open source Plone content management system.
A new version of our authoring technology for adaptive learning is now available.
CAPE 2.7 focuses on support for larger-scale design families and courseware product lines.
CAPE has long supported instructional design patterns as a means of capturing commonalities among sets of
courseware designs.
These patterns are represented in CAPE as abstract designs, or models, that can be instanced in various ways.
At the same time, CAPE's data modeling facility, Condition Sets, and adaptive content features have been used
to create designs that are data-driven, providing an alternative abstraction facility.
The power of using these capabilities in concert has been demonstrated by a
web-based authoring system for a class
of engineering homework problems that provide students with adaptive remediation.
CAPE 2.7 extends these capabilities with new features that support more agressive design reuse strategies.
An example of such a strategy is the ACT Online
courseware product line.
This training system, sponsored by the U.S Federal Emergency Management Agency, provides 9 courses on cyber-terrorism
incorporating a total of 45-50 modules that share a common instructional design macro-structure called STAR Legacy.
The
TRUST Academy Online (TAO) Portal supports online community outreach
for the NSF
TRUST Science and Technology Center. TAO reuses the dissemination portal framework
that we originally developed for the NSF VaNTH Engineering Research Center for Bioengineering Educational Technologies.
Like the VaNTH Portal, TAO provides "courseware profiles" that bundle sets of learning materials
with descriptions, metadata, and scaffolding resources.
For TAO, we have added a new type of content object, called "project profiles," that
tell the stories of TRUST research projects and provide access to resources
produced by the projects such as papers, presentations, and posters.
Project designers Laszlo Juracz and Gabor Pap have recently refined this Plone-based framework to provide
a fresh "look and feel" featuring visual browsers for content objects along with other usability, content, and interoperability enhancements.
This further evolution of our dissemination technology reflects a continuing commitment to provide educators
and other users access to materials and resources produced by our research partners through the World Wide Web.
Following the introduction of powerful search engines, searching joined hyperlink-based browsing to form the two dominant
use metaphors of the World Wide Web.
Online learning designs, even when constructed as browsable hypermedia, seldom employ searching as an important element of the learning experience.
With CAPE 2.7, we have provided infrastructure that allows authors to incorporate this capability into their learning designs.
Our first use of this infrastructure is in the ACT Online program, which provides course modules based on the STAR Legacy
instructional design pattern. This inquiry cycle provides sets of learning resources and self-assessment questions
anchored on an overarching challenge.
The ACT Online search facility indexes these resources and questions so that they can be searched on demand by learners.
The user interface of this facility, shown above, supports a "live search", where keywords and phrases are matched and returned as
suggestions to assist the search.
The interface returns search results on separated "Resources" and "Questions" tabs sorted by relevance with direct navigation to the content objects.
The new infrastructure provides services for indexing and searching
that are accessed through the existing eLMS Platform Services at delivery time.
Indexing services build up search indices and
search services then support assisted retrieval by keywords and phrases.
The search infrastructure reflects our ongoing efforts to make online learning more indigenous to the web.
The Adaptive Cybersecurity Training (ACT) Online program, sponsored by the U. S. Department of Homeland Security,
is using CAPE to create a
courseware product-line: 70-80 modules comprising 9 courses in 3 tracks with progressing levels of difficulty.
Some innovations arising from this program are directly influencing features of CAPE.
The ACT Online approach to product line development with CAPE involves extensive use of instructional design patterns.
These patterns (basically, modeling abstractions) are instanced using sets of specifications
for parts such as learning resources, formative and summative assessments, and scaffolding elements.
The patterns govern the integration of these parts when forming modules and complete courses,
as well as artifacts that support specialized role-players in the development process (the production line.)
One ACT Online-inspired feature coming in CAPE 2.7 involves a new method of incorporating
these part specifications into designs.
The specifications ultimately reside in CAPE's data modeling facility known as condition sets.
The new feature permits the specifications to permanently reside on the file system in the development environment
where they can be managed using tools like CVS or Subversion.
Condition sets will contain a new modeling element, called ImportFile, that incorporates
such files into condition sets at courseware packaging time.
Another area where ACT Online is innovating aggressively concerns custom interfaces and templating.
We will hightlight some influences of these innovations on CAPE and eLMS technology
in future news.
CAPE's online documentation has been indexed by the Google search engine and can now be searched from the documentation page.
A new search element has been added to the documentation page, accessed using the link on the sidebar at left, that performs
a search restricted to the documentation area on the CAPE server.
Searchability is a documentation feature that has been requested by several users.
Rather than address this issue by building our own indices and server-based search tool, we have elected (like many others) to use
features provided by web search engine providers.
An available update to the CAPE authoring environment ends its long, intimate relationship with
Microsoft's Internet Explorer browser.
Perhaps that's a tad strong—the relationship is no longer exclusive.
Since the beginning, CAPE has used IE for all of its previewing services:
for locally viewing adaptive content and design elements, as well as for previewing
in-progress courseware designs.
This exclusivity has extended to browser-based HTML editing with the recently incorporated TinyMCE editor.
But more and more authors are telling us that they no longer use IE as their default browser, and
we have responded by honoring their choice.
With around 80% marketshare, IE is certainly not a browser to be neglected.
But it presents security issues (its support for ActiveX controls through scripting, for example) that other browsers simply don't face,
particularly when accessing local files.
After many exploits, IE7 addresses these issues with a hardened security policy that is not easily (nor safely) softened for a particular trusted application.
These restrictions present certain inconveniences to authors using CAPE's previewing services.
The update adds a new setting to the
CAPE Settings component that allows authors to select whether they wish to
use their default browser or Internet Explorer for previewing within CAPE.
A further advantage of this setting is that—when IE is not the author's default browser—content rendering between browsers
can be easily checked by switching this setting without the need to change default browser.
Version 2.6 of the CAPE authoring environment is now available for download and installation.
The installer can be accessed by selecting the
Release Notes item on the menubar at left.
Detailed information about what is new in this release, beyond that provided in the article below,
is included on that page.
CAPE is not available directly to the public, but rather by agreement with the VaNTH ERC or through
affiliated institutions, centers, or projects.
Inquiries about using CAPE (and eLMS) in other contexts can be made using contact information available
on this site.
For new users, especially those participating in the VaNTH Workshop series, there is an available
Installation Guide that provides
detailed instructions, demonstrational resources, and access to a tutorial CAPE project.
A forthcoming major version of the CAPE authoring environment provides extensions to the design language, particularly in the area of assessments,
improves scaffolds for the author, and includes a new courseware package export facility.
CAPE's
online documentation
has been given a fresh "look",
and bundled with this release is a new version of the
Generic Modeling Environment (GME) infrastructure.
Multiple Choice assessment items can now include an "Other" (or, arbitrary) choice.
An Image Multiple Choice item is introduced specifically for these kinds of questions.
A new File item is added that allows learners to upload a file as the response.
Other design language additions include Comment attributes for adaptive elements such as
Action, Condition, and Select.
Several more attributes can now be accessed via author-defined external editors.
The design environment adds features that further contribute to usability.
For example, when an Assessment element is created in a courseware model,
a corresponding Outcomes element is automatically created and associated.
Changes to the Assessment definition are now dynamically reflected in the Outcomes.
Similar conveniences are added when defining Condition Sets, for example, coordinating the naming
and definition of Functions.
The new GME version adds a model window navigation toolbar with capabilities similar to the history
features of web browsers.
We anticipate the release of CAPE 2.6 within a week.
The
Software Technology for Action and Reflection (STAR) Legacy Cycle was originally developed by the
Cognition and Technology Group at Vanderbilt as a "software shell" for
anchored instruction.
This active learning style grounds the learning process on authentic problem situations (challenges) and
further promotes the development of learning skills by allowing learners to explore a multiplicity of learning
resources and providing activities that facilitate the learner's self-assessment in guiding this exploration.
CAPE can be used to author online learning experiences in this style and
has long supported a design pattern for the so-called "six-phase" Legacy Cycle.
For the NSF TRUST STC, we have developed additional scaffolding for this
pattern that supports familiarizing learners with the cycle and provides content templates
for the individual cycle phases.
The new DHS ACT Online Program (described below) will use another manifestation of Legacy called
the five-phase cycle (shown at right above), employed successfully for adult online continuing education by
the
IRIS Center at Vanderbilt.
In pursuing this pedagogical approach, we will be extending CAPE and eLMS to combine design
templates and delivery interfaces.
This represents a departure from our fundamental strategy of adaptive design with CAPE and a consistent
presentation of design enactments with eLMS.
The extensions will open many new opportunities for supporting particular learning styles.
CAPE and eLMS will be used as authoring and delivery technologies for a new cybersecurity training grant funded by
the U.S. Department of Homeland Security.
The Adaptive Cybersecurity Training (ACT) Online program will provide multi-level, multi-track training
for first responders in the critical area of information assurance.
The three-year, $4M program will be lead by Prof. Dipankar Dasgupta of the University of Memphis.
The online aspects of the program will be developed in collaboration with
Vanderbilt University's Institute for Software Integrated Systems (ISIS).
Adaptive features of CAPE and eLMS will be used to customize the training to reflect
the prior knowledge and skills of individual learners and to facilitate their progress through various modules
culminating in the granting of certifications.
A new version of the CAPE authoring environment is available today as an automatic update.
The principal new feature in this release is an integration of
the
TinyMCE
web-based HTML editor
(described in an article below).
TinyMCE can be configured as an external HTML editor using the
CAPE Settings component.
We were so pleased with the integration of this editor into our web-based authoring interfaces,
we thought we would also offer it as an option to CAPE authors.
A new beta release of the
Parallels Desktop for Mac virtualization solution
makes significant strides towards easing the use of Windows applications, like the CAPE authoring environment, on Mac OS X.
A new mode, called Coherence, allows Windows applications to appear directly on the Mac desktop, enabling easy switching between
applications on the two operating systems.
This feature, along with drag and drop support for file sharing and better integration with Apple's dual-boot solution, called Boot Camp,
make using CAPE on Macs easier than ever before.
Larry Howard, the principal designer of CAPE (and Mac fan), uses Parallels Desktop on his MacBook Pro extensively for CAPE
courseware design. “This environment is excellent for checking web content among a number of different browsers”, he said.
We expect these new features to be just the beginning of powerful capabilities for seamlessly running applications from multiple operating systems enabled
by virtualization technology.
We further anticipate that competition among solution providers in this sector will bring these capabilities quickly to market.
To improve the ease of use of new eLMS authoring and adaptation tools, we have added support for
a full-featured open source Javascript HTML editor called
TinyMCE.
This enhancement allows assessment and content elements to be edited in a WYSIWYG (
what you see is what you get) manner
to assist users with limited knowledge
of HTML.
TinyMCE is an extensible, cross-platform HTML editor and among the extensions we have provided
are support for data-driven resources and dynamic content.
Resources uploaded through the authoring interface can be inserted into HTML content and manipuated by TinyMCE,
including support for resource browsing that ensures resources are properly referenced.
Dynamic content elements in TinyMCE appear in their substituted form and an editing interface supports defining, altering, and
evaluating the corresponding content expressions in the facility known as <py> tags.
This interface further supports referencing data items through an integrated data space browser.
With these extensions, TinyMCE-based HTML editing for web authoring in the delivery environment mirrors usability features
supported by CAPE itself.
CAPE is a state-of-the-art visual language authoring environment for adaptive online learning, and
its ability to be used to design new web authoring tools for eLMS provides a glimpse of its power and versatility.
The first such tool (shown at right cycling through its views) supports many of the capabilities of CAPE
itself—dynamic content, integrated formative assessments,
adaptive (potentially progressive) feedback, instructor interventions, and data definitions (including derived data and functions).
That these varied design specifications, along with file-based resources, can now all be dynamically created
within the delivery environment
is an achivement that enables multiple forms of
assignment-time adaptation and just-in-time authoring.
The demonstration we have chosen for these capabilities is a kind of Socratic tutor, where exercises proceed
by asking learners questions, adaptively providing feedback based on their responses, and then adaptively selecting follow-on questions.
The web-based authoring tool is based on a generalized version of a design pattern by Prof. Robert Roselli of Vanderbilt University referenced in an earlier
article below. The authoring tool was created in collaboration with Prof. Roselli and two bioengineering graduate students: Kathryn Dwyer and
Andrew Slatton.
The authoring tool is directly contained within a courseware implementing the Socratic tutoring design pattern and,
like CAPE itself, provides an integrated debugger.
Communication between the authoring tool and the delivery component is via the new
assignment profile.
Instructors interested in this new courseware can contact
Prof. Roselli
and inquiries about the new features can be directed to us.
To complete our scaffolding for new late authoring capabilities within the delivery environment,
CAPE adds support for representing file-based resources using its dynamic data representation
(
Condition Sets) and eLMS adds support for uploading and using such resources at delivery time.
In addition to late authoring, these capabilities can also support learners uploading file-based
resources created as workproducts within an online activity.
CAPE authors identify file-based resources as part of the specification of a learning design.
When a CAPE design is uploaded to an eLMS server for delivery, these file-based resources are automatically gathered and incorporated into
a package sent to the server.
As the courseware design is enacted with learners, these resources are made available to satisfy requests coming from the learner's browser.
This approach to resource management is intrinsic to the designs of CAPE and eLMS, and it creates a limitation that
all file-based resources must be identified at design time.
Data-driven resources remove this limitation by enabling an alternative approach: representing
resources as
data.
CAPE supports this alternative through a kind of condition called
Base64 that can contain encoded file resources.
eLMS now supports uploading and encoding such resources using a new service of the delivery engine, called
uploadConditions,
and supports referencing such conditions through
Dynamic resources defined at design time, called
gateways.
A mini-project called
Data-Driven Resources in the
CAPE author area of the Repository documents and demonstrates
these capabilities.
Introduced back in CAPE 2.3,
data-driven assessments began to blur the line between designs at authoring- and delivery-time.
The innovation was that an assessment's design could be contained in
CAPE's dynamic data representation, called
Condition Sets, thereby enabling it
to be modified during courseware delivery to add, remove, or alter
the kind of questions presented by the assessment.
Prof. Robert Roselli of Vanderbilt used this capability to create a highly generative design pattern in CAPE that
supports authoring using only data structures contained in condition sets (detailed in this
paper).
The design pattern, involving adaptive sequencing and adaptive content in addition to data-driven assessments, acts (essentially) as a special-purpose delivery engine that enacts the "instructions" defined
in the data structures, including what questions to ask learners and how to react to their responses.
Similar techniques were used by Prof. Eric Perreault of Northwestern to create a lab preparation exercise
where questions are randomly drawn from mutliple question banks with progressively increasing levels of difficulty.
Data-driven Condition Sets represents our next step along this path,
enabling data contained in condition sets to define other condition sets that can be initialized during courseware delivery.
Whereas Actions and eLMS web services enable altering a courseware's dynamic state, data-driven condition sets
focus on defining (or re-defining) this state.
The definitional capabilities extend to CAPE's computational features (Derived conditions and Functions)
that heretofore could only be defined at authoring-time.
A supporting mini-project, called "Data-Driven Condition Sets", is provided in the CAPE author area of the Repository.
We are introducing data-driven condition sets as scaffolding for forthcoming capabilities that will address assignment-time adaptation and
authoring.
These capabilities will allow CAPE-authored courseware designs to be adapted by educators through the web using authoring interfaces designed
with CAPE that are adaptively integrated directly into the courseware.
The effect will be to enable a spectrum of authoring times, from early design within the authoring environment
to late design within the delivery environment, and authoring tasks, from design creation to design adaptation.
The CAPE Wizard (shown at right) is an authoring environment extension that supports creating new design elements based on templates.
The initial version of this extension—introduced in CAPE 2.5—supported a set of "built-in" templates.
After it was introduced, authors inquired about using the Wizard to support their own templates.
A CAPE update released today provides this capability.
The Repository Browser CAPE component now supports designating a model in the Repository as a template.
Only those models that are self-contained (that is, have no external dependencies by reference or derivation) can be templates.
Use the Mark as template choice from the context menu (right click) to designate a template.
Authors can (and should) provide a short description of the template that will be displayed by the CAPE Wizard.
Use the Edit template description choice from the context menu to modify the description at a later time.
Authors are further encouraged to use model annotations to assist other designers in reusing templates.
Each time the CAPE Wizard is started, the Repository is queried for available templates.
When selecting a kind of element in the Wizard, if templates are available, then an associated radio button on the
interface enables switching between built-in and repository templates.
When the latter is selected, templates are presented in a listing view for selection and creation.
A series of four workshops will be delivered by the
VaNTH ERC over the
coming academic year that are based on methods and technologies researched by the center.
The workshops will include training
on instructional design using the
How People Learn
(HPL) framework, as well as the use of the CAPE authoring environment for designing online and blended learning environments.
The two-day workshops will be conducted by Dr. Alene Harris, the director of VaNTH's Education Program, and
Dr. Stacy Klein, an education and bioengineering faculty participant in VaNTH—both highly experienced
workshop facilitators.
The location will be Vanderbilt University in Nashville, TN.
The workshops will include contributions by other VaNTH participants from the ERC's
Bioengineering Domains and Learning Technology Thrusts, including Larry Howard, the principal
designer of CAPE. The workshops will include a half-day focusing exclusively on introducing CAPE and a half-day synthesis
of pedagogy and technology-focused aspects.
Those interested in participating in one the workshops can help determine the dates by
pre-registering online by
October 2, 2006. Additional information is available at the
workshop site,
including details of scholarships and stipends that have been established to help extend participation.
Updated: October 11, 2006
The following four 2007 dates have been selected for the VaNTH workshops:
January 8-9,
February 5-6,
April 16-17, and
May 7-8.
You can register for one of these workshop dates and apply for the available scholarships
online.
eLMS provides a capability, called
Profiles, that can be used to support various kinds of online collaborative work.
Class profiles are shared by the students in a class and Courseware profiles are shared between all deliveries of a particular courseware.
Such "global" state enables information sharing among students or between students and the instructor (or teaching assistants) of an eLMS class.
To facilitate the use of these capabilities by CAPE authors, we have produced a set of
demonstration designs made available in the Repository.
The designs are annotated in CAPE and web pages further document them and support trying them out.
Dialoging demonstrates using the Class profile to support
dialogs between individual students and the instructor of a class.
Such dialogs enable instructors to iteratively give feedback to a learner during the course of an activity performed online
and allow the learner to respond to this feedback.
Research Designs demonstrates support for
experimental designs where learners can be randomly assigned to groups to adaptively
receive different "treatments".
Discussions demonstrates forums where participation
can be class-wide or in the context of randomly formed mini-groups.
The importance of these capabilities for eLMS courseware is that collaborative features can
be situated within courseware authored with CAPE.
Adaptive design capabilities supported by CAPE can be used to scaffold the use of these kinds of elements
or to integrate them into designs that also provide features for individualized instruction.
eLMS has introduced a new courseware delivery mechanism, called
Quota Links,
that allows learners to access eLMS courseware using special limited use URLs.
This new mechanism greatly extends our existing integration solutions by not requiring
the use of eLMS, or a cooperating learning platform, to manage learner identities and class rosters.
When learners access a Quota Link, a panel (shown at right) is presented that collects their
identity information and then launches a courseware delivery for them.
This approach is different from existing forms of anonymous delivery supported by eLMS in that
it allows learners to continue or review an eLMS courseware over any number of browser sessions.
Learners choose their own passcode that, together with their email address, uniquely identifies their
courseware delivery record.
Quota Links permit only a fixed number of courseware deliveries (the quota) to be started.
The eLMS Author interface supports the creation of Quota Links.
The delivery records resulting from the use of such links can be retrieved in a similar way
as those from SCORM packages.
A new version of the CAPE authoring environment has been released and can be downloaded
through the
release notes page.
CAPE 2.5.1 is an incremental release of the authoring environment that provides bug fixes and a couple
of significant enhancements.
The first of these is a new project launcher (shown at right)
that reflects our continuing efforts to customize and brand the Generic Modeling Environment (GME)
infrastructure upon which CAPE is built.
This dialog is specialized for the CAPE project file type (.cape)—introduced in CAPE 2.5—and
can be used to open existing CAPE projects, create new projects, or
select from the recently opened project list maintained by GME.
The more significant enhancement in this release is the introduction of an automatic update feature.
This feature can be configured using the CAPE Settings component
to check for updates each time CAPE is launched or on a fixed schedule.
With this new component, we will be able to distribute changes to the authoring environment through the web in the future
without the need to provide incremental releases like this one: easier for you, easier for us.
Virtualization solutions are appearing that allow Windows applications,
including the CAPE authoring environment, to run on Intel-based Macs.
One currently available solution is
Parallels Desktop.
Virtualization performs much better than emulation solutions that have been available in the past,
such as
VirtualPC,
and are easier to use than dual-boot solutions like Apple's
BootCamp.
We have confirmed that CAPE runs quite well in Parallels Desktop.
To use this solution, users must purchase a low-cost Parallels license and supply a
Microsoft Windows operating system.
Updated: August 12, 2006
Microsoft has
announced that they will not carry forward VirtualPC for Intel-based Macs, so it seems that virtualization solutions
and dual-booting will be the only practical approaches to running CAPE on Macs.
VMware has also announced a new virtualization solution, and we will be tracking the development of this product.
A new version of the CAPE authoring environment has been released and can be downloaded
through the
release notes page.
CAPE 2.5 provides improvements to its support for interoperability, changes to its
metadata representation, and performance and functionality enhancements to the CAPE Repository.
Interoperability improvements address three areas: multiple external content editors,
SCORM/IMS content packaging, publishing to a dissemination portal.
GME's support for defining an exernal editor has been extended to define
different editors for attributes of CAPE objects containing text, HTML, and Python.
The CAPE Settings component adds support for these settings.
SCORM packaging, introduced by eLMS in 2.4, can be performed directly from CAPE 2.5.
Support for IMS content packaging is added for certain legacy CMS that do not support SCORM.
A new VaNTH dissemination portal, based on the Plone
content management system, has been developed and support is added for directly publishing information about courseware to the
portal from CAPE 2.5. This metadata-based capability will also be used by TRUST authors to
publish to the forthcoming TRUST Academy Online (TAO) portal.
CAPE's metadata support has undergone its first revision since the original definition of
the VaNTH paradigm.
The changes reflect a restructuring to support more systems of description.
Metadata specifications now consist of three kinds:
Tags,
MultiTags, and
CompoundTags.
Tags support defining a single value, provide a description for the user, and can define a validation predicate.
MultiTags support similar definitions, but serve as containers for multiple values.
CompoundTags provide a (potentially recursive) container for Tags and MultiTags to
support a set of specifications employed as a unit.
Metadata folders, which can contain each of the above kinds of objects,
are now recursive to support hierarchically organized collections.
The CAPE Repository incorporates improvements that address scalability and evolution.
Launch time is dramatically decreased using a lazy browsing strategy and interchange performance
is improved using a new binary format.
Support has been added for search and retrieval, with server-based indexing.
Repository assets created with earlier versions of CAPE can now be loaded into newer CAPE projects with
automatic version migration.
The Notes Editor of the eLMS courseware delivery interface now supports HTML-based rich text editing,
including support for hyperlinks.
This upgrade to the previous plain text facility is based on the
obedit Flash component
created by Richard Podsada at
oblius.com.
As a service to the GME community, we are providing a component generation framework called
PyGME for GME components
written in Python.
A new version of the CAPE authoring environment has been released and can be downloaded
through the
release notes page.
The next version of the CAPE authoring environment will provide
the first installment of "instructor-in-the-loop" capabilities called
interventions. Since CAPE/eLMS courseware are predominantly used in
blended learning environments, these capabilities target coordination
between in-class and outside-class elements of some overarching learning design.
The first capabilities will concern modeling synchronization points at which a
courseware design will suspend awaiting action by an Instructor or teaching
assistant. Release of learners from these synchronization points can be
selective or collective within the context of an eLMS class.
Other new features of CAPE 2.4 include exception sequencing
edges in courseware models and dynamic Resources models. The former
pertains to the Action, Condition, and Select modeling
elements. The design language now supports defining delivery pathways that
are followed in the event that the execution of statements or evaluation of
expressions raises an exception. Resources models can now be
associated with a folder on the local file system, which can be filtered for
particular content types. Files in such folders are automatically
incorporated into content packages when previewing or uploading courseware.
The next version of eLMS will support packaging and delivery of
courseware authored with CAPE by a SCORM-compliant LMS. An eLMS server
will still be required to enact the courseware design and serve learning
content. This interoperability solution wraps the eLMS courseware as a
single SCORM Shareable Content Object (SCO) that can be assigned, launched, and
resumed from a SCORM LMS, with outcomes returned to the LMS. Complete
delivery records can be accessed from eLMS by the user that generated the SCORM
package using a data mining interface similar to that provided to eLMS
Instructors. These capabilities replace our announced objective to develop a
PowerLink integration with WebCT.
This update to our integration with the Blackboard Learning
System supports writing learner outcomes from eLMS
courseware back to a Blackboard gradebook. The gradebook recording
feature can be enabled or disabled by system administrators at building block
installation time.
These mailing lists support CAPE courseware authors and eLMS
users, including those that access eLMS through the Blackboard Learning System.
Subscribe to the CAPE list
here
and the eLMS list
here.
